2025/12/05 by Pascal O. Title, L. Francisco Henao‐Díaz, Rosana Zenil‐Ferguson +1 · 1 voice · 3 citations
Biochemistry, Genetics and Molecular Biology · Earth and Planetary Sciences · #Diversification (marketing strategy) #Evolution and Genetic Dynamics #Evolution and Paleontology Studies #Extinction (optical mineralogy) #Genetic algorithm #Genomics and Phylogenetic Studies #Lineage (genetic) #Macroevolution #Pace #Selection (genetic algorithm) #Species richness
paper · pdf · doi:10.1093/sysbio/syaf086
published in Systematic Biology 75(3), 391-405 (Oxford University Press)
openalex created_date 2025/12/05 · openalex publication_date 2025/12/05 · openalex updated_date 2026/07/26
Systematists have long been fascinated by the astounding variation in species diversity across the various branches of the tree of life, a net result of the uneven rates at which lineages undergo speciation and extinction over time. The past 30 years have seen the development and widespread application of tools that allow diversification to be quantified and characterized in empirical data sets. These advances have, in turn, enabled the statistical evaluation of hypotheses about the causes behind the uneven distribution of species richness among lineages, leading to a more nuanced understanding of diversification rate variation, as reflected in an ever-expanding literature. Here, we provide a brief review of the current understanding of these models, the types of questions they address, and some of their collective limitations, with a focus on tree-based analyses of reconstructed phylogenies. Based on this overview, we outline future considerations in the lineage diversification research program, including the potential for machine learning to revolutionize the field by making model selection and parameter estimation more efficient in highly complex models. We interpret the recent slowdown in publication pace as a sign of a maturing field, where systematists are taking a step back after becoming better equipped to understand the technicalities and current limitations of these methods, leading to more careful applications and a greater embrace of uncertainty.